Human liver methenyltetrahydrofolate synthetase: improved purification and increased affinity for folate polyglutamate substrates

154

Biochimica et Biophysica Acta 911 (1987) 154-161 Elsevier

BBA 32695

H u m a n liver m e t h e n y l t e t r a h y d r o f o l a t e s y n t h e t a s e : i m p r o v e d purification and i n c r e a s e d affinity f o r f o l a t e p o l y g l u t a m a t e s u b s t r a t e s R i c h a r d B e r t r a n d a, R o b e r t E. M a c K e n z i e b a n d J a c q u e s Jolivet a a [nstitut du Cancer de Montreal, Montreal (Canada) and b Department of Biochemistry, Faculty of Medicine, McGill University, Montreal (Canada) (Received 29 July 1986)

Key words: Methenyltetrahydrofolate synthetase; Inhibitor; Kinetic constant; (Human liver)

Methenyltetrahydrofolate synthetase (5-formyltetrahydrofolate cyclodehydrase (cyclo-ligase) (ADP-forming) EC 6.3.3.2) catalyzes the ATP- and Mg 2+-dependent transformation of 5-formyltetrahydrofolate (leucovorin)

to 5,10-metbenyltetrahydrofolate. The enzyme has been purified 49000-fold from human liver by a two-column procedure with Blue Sepharose followed by folinate-Sepharose chromatography. It appears as a single band both on SDS-polyacrylamide gel electrophoresis ( M r 27000) and on isoelectric focusing (pI = 7.0) and is monomeric, with a molecular weight of 27000 on gel filtration. Initial-velocity studies suggest that the enzyme catalyzes a sequential mechanism and at 30°C and pH 6.0 the turnover number is 1000 min -1. The enzyme has a higher affinity for its pentaglutamate substrate ( K m ~- 0.6/*M) than for the monoglutamate (K m = 2 pM). The antifolate methotrexate has no inhibitory effect at concentrations up to 350 /,M, while methotrexate pentaglutamate is a competitive inhibitor with a K i = 15 p M . Similarly, dihydrofolate monoglutamate is a weak inhibitor with a K i = 50/*M, while the pentaglutamate is a potent competitive inhibitor with a K i of 3.8/,M. Thus, dihydrofolate and methotrexate pentaglutamates could regulate enzyme activity and help explain why leucovorin fails to rescue cells from high concentrations of methotrexate. Introduction Methenyltetrahydrofolate synthetase (5-formyltetrahydrofolate cyclodehydrase (cyclo-ligase) (ADP-forming), EC 6.3.3.2) catalyzes the transformation of 5-formyltetrahydrofolate (5-HCOH4PteGlu; folinic acid; leucovorin; citrovorum factor) to 5,10-methenyltetrahydrofolate (5,10CH +-H 4PteGlu) in the presence of ATP and Mg 2+

Abbreviations; Pipes, 1,4-piperazinediethanesulfonic acid; Hepes, 4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid; Mes, 4-morpholineethanesulfonic acid. Correspondence: Dr. J. Jolivet, Institut du Cancer de Montr6al, Centre Hospitalier Notre-Dame, 1560 Sherbrooke est, Montr6al H2L 4M1, Canada.

(reaction I). 5-HCO-H4PteGlu+ATP +ADP+P i

Mg 2+ ---, 5,10-CH+-H4PteGlu (1)

Greenberg [1] was the first to show that 5HCO-H4PteGlu was converted enzymatically to another folate involved in purine biosynthesis. Sheep liver enzyme activity was later partially purified by Peters and Greenberg [2-4] and the enzyme was more recently highly purified and characterized from Lactobacillus casei [5] and rabbit liver [6]. In addition to methenyltetrahydrofolate synthetase, 5-HCO-H4PteGlu is also a substrate for N-formyl-L-glutamate:tetrahydrofolate formyltransferase (EC 2.1.2.6). Bortoluzzi and

0167-4838/87/$03.50 © 1987 Elsevier Science Publishers B.V. (Biomedical Division)

155 MacKenzie [7], however, have shown in pig liver extracts that this enzymatic activity is a less specific activity of N-forminino-L-glutamate:tetrahydrofolate formininotransferase (EC 2.1.2.5) with a 2000 : 1 activity ratio in favor of the latter. Consequently, this 5-HCO-H4PteGlu metabolic pathway is probably of little if any physiological significance. 5-HCO-H4PteGlu is administered clinically either as a rescue agent following high-dose antifolate therapy [8] or more recently in association with the antimetabolite 5-fluorouracil to overcome drug resistance due to insufficient intracellular folate concentrations [9]. It is thus important to understand better the human methenyltetrahydrofolate synthetase, since 5-HCO-H4PteGlu has no known folate cofactor activity per se and must first be converted by the enzyme to 5,10CH+-H4PteGIu to be able to replenish intracellular folate pools. In the present study, we have obtained and characterized highly purified human liver methenyltetrahydrofolate synthetase and observed that it has a higher affinity for folate and antifolate polyglutamate substrates. Materials and Methods

Chemicals. Folic acid (PteGlu), tetrahydrofolate (H4PteGlu) and (6R,S)-5-methyltetrahydrofolate (5-CH3-HaPteGlu) were purchased from the Sigma Chemical Co. (St. Louis, MO). Methotrexate (4NH2-10-CH3-PteGlu ) and ( 6 R , S ) - 5 - H C O H4PteGlu (calcium salt) were obtained respectively from the Drug Synthesis and Chemistry Branch, National Cancer Institute (Bethesda, MD), and from the Aldrich Chemical Co. (Milwaukee, WI), and purified by DEAE-cellulose chromatography (Whatman, U.K.) with a linear gradient of 0.01-1.5 M NH4HCO 3 [10]. (6R,S)-5-HCOHaPteGlu, potassium salt, was made from the calcium salt by exposure to potassium oxalate and the calcium oxalate formed was removed by centrifugation. Purified synthetic methotrexateGlu 5 and PteGlu 5 (methotrexate and PteGlu containing a total of five glutamine residues) were provided by Dr. C.M. Baugh (Department of Biochemistry, University of South Alabama, Mobile, AL) and Dr. Carmen J. Allegra (National Cancer

Institute, Bethesda, MD). (6S)-5-HCO-H4PteGlu . was prepared as follows: PteGlu, was first reduced to H2PteGlu . by sodium hydrosulfite [11] following which HzPteGlu . was further reduced to (6S)-H4PteGhi n using an NADPH-generating system [12] with partially purified Lactobacillus casei dihydrofolate reductase (EC 1.5.1.3) obtained from the New England Enzyme Center (Boston, MA). (6S)-H4PteGlu . was converted to (6S)-5-HCO-H4PteGlu . by carbodiimide-induced formylation as described by Moran and Colman [13]. (6S)-5-HCO-H4PteGlu . was then purified by a DEAE-cellulose column as described above. 5,10-CH+-H4PteGlu and 10-HCO-H4PteGlu were prepared as described by Rabinowitz [14]. Aminohexyl-Sepharose-4B, Blue Sepharose CL-6B, Sephadex G-25 and G-200 were obtained from Pharmacia Fine Chemicals (Uppsala, Sweden). 1Ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC), polyoxyethylene sorbitan monolaurate (Tween 20), bovine albumin factor V, D-glucose, o-glucose 6-phosphate and acid-washed activated charcoal were purchased from Sigma. Urea and /%mercaptoethanol were purchased from J.T. Baker chemical Co. (Phillibsburg, N J) and glycerol from BDH Chemical Co. (Toronto, Ont.) Pipes, Mes, Hepes, Tris, ATP, ADP, NADP and its ~educed form (NADPH), D-hexose-6-phosphotransferase (EC 2.7.1.1) and glucose-6-phosphate dehydrogenase (EC 1.1.1.49) were purchased from Boehringer-Mannheim (Penzberg, F.R.G.). Electrophoresis-grade reagents and protein dyebinding reagent were obtained from Bio-Rad (Richmond, CA). All other chemicals were of reagent grade and were purchased from Fisher Scientific Co. (Fairlawn, N J). 5-HCO-H4PteGlu-aminohexyl-Sepharose. The affinity column was prepared by a modification of a method previously described for the synthesis of methotrexate-Sepharose [15]: (6R,S)-5-HCOH4PteGlu (50 mg) was dissolved in 20 ml of distilled deionized water and added to a suspension of activated (pH 4.5) AH-Sepharose-4B (5 g). The mixture was stirred for 30 min at 25 ° C before adding 1-ethyl-3-(3-dimethylaminopropyl)carbodiimde (250 mg) and the pH was adjusted to 6.0. The final mixture was stirred for an additional 24 h. The gel was washed and stored at 4°C in 50 mM Pipes, 50 mM/3-mercaptoethanol (pH 6.0).

156

Assay system. Methenyltetrahydrofolate synthetase activity was measured by monitoring the increase in absorbance at 360 nm due to the formation of 5,10-CH+-H4PteGlu [16] using a 'Spectronic-2000' spectrophotometer (Bausch and Lomb, Rochester, NY). Reactions were carried out in quartz cuvettes of either 1 cm or 10 cm optical path-length and the temperature was maintained at 30°C using a water-jacketed sample compartment. The 1 cm cuvette was used to detect enzyme activity during purification. The kinetic measurements were initially performed using the 1 cm cuvette. Measurements at low folate concentration were taken using the 10 cm cuvettes. The assay mixture contained 50 mM Mes, 10 mM fl-mercaptoethanol, 10 mM magnesium acetate, 0.5 mM ATP and 0.2 mM (6R,S)-5-HCOH4PteGlu (calcium salt) (pH 6.0). One unit of activity represents 1 /~mol 5,10-CH+-H4PteGIu formed per min (E360 = 25.1 • 103 M -1 • cm -1) [16]. The velocity is expressed as /~mol-min -1. mg -1. Enzyme activity under varying pH conditions was determined at 30°C using different buffers depending on their pK a. The enzyme assay was modified to account for the spontaneous formation of 10-HCO-H4PteGlu from 5,10-CH ÷H4PteGlu seen at higher pH values. The reaction was allowed to proceed for 1 h in 0.1 M buffer containing 10 mM /3-mercaptoethanol. Substrate concentrations used were 2 mM ATP, 20 mM magnesium acetate and 1 mM (6R, S)-5-HCOH4PteGlu to ensure that saturated levels of substrates were maintained over the entire pH range. The reaction was stopped by adding citric acid saturated with ammonium sulfate to a final concentration of 0.2 M at pH 3.5. Assay mixtures were then treated in a boiling water bath for 30 s and chilled on ice. 5,10-CH+-H4PteGlu concentrations were calculated from absorbance at 350 nm after centrifugation. Purification of methenyltetrahydrofolate synthetase. Human liver was obtained at autopsy and kept at - 8 0 ° C . Approximately 40 g were homogenized in 100 ml of 20 mM Pipes, 20 mM /3-mercaptoethanol, 10 mM magnesium acetate at pH 7.0 (buffer 1) and the homogenate was centrifuged at 12000 x g for 30 min. The supernatant was saturated at 70% with solid (NH4)2SO 4 (43.6 g/100 ml) and the mixture was stirred for 30 rain.

After centrifugation (12 000 x g, 30 min) the precipitate was resuspended in a minimal volume (20 ml) of buffer 1 (pH 7.0) and desalted by centrifugation (5000 x g, 10 min) through a 3 x 8 cm column of Sephadex G-25. All steps were performed at 4°C. The fraction containing protein was then applied to a Blue Sepharose column (1.6 x 100 cm) which had been equilibrated at 4°C with buffer 1 at pH 7.0. The column was then washed with 4 liters of buffer 1 containing 0.1% (v/v) Tween 20, 10% (v/v) glycerol (buffer 2) and 0.35 M KC1 at p H 7.0 (flow rate: 120 ml/h). Methenyltetrahydrofolate synthetase was eluted with 500 ml buffer 2 in 0.35 M KC1 containing 0.1 mM (6R,S)-5-HCO-H4PteGlu (calcium salt). Fractions containing activity were pooled and the solution was dialysed twice against 24 liters of buffer 2 containing 24 g acid-washed activated charcoal to remove 5-HCO-H4PteGlu prior to the next step. To the dialysed solution, 0.15 M KC1 was added and the pH was adjusted to 6.5. The solution was applied to a column (1 x 30 cm) of 5-HCOH4PteGlu-aminohexyl-Sepharose equilibrated at 4°C with buffer 2 containing 0.15 M KC1 at pH 6.5. The column was then washed with 1 liter of the same buffer and elution of the enzyme was accomplished by applying 1 mM ( 6 R , S ) - 5 - H C O H4PteGlu potassium salt. The enzyme was eluted as a sharp band and dialysis was performed against charcoal as described above prior to kinetic analysis.

Gel eleetrophoresis, gel filtration and protein assay. Purity and molecular weight of the enzyme preparations were determined by SDS-polyacrylamide gel electrohporesis according to the method of Laemmli [17] in 11% vertical slab gels. Gels were stained with silver nitrate according to the method of Morrissey [18]. The isoelectric point of the purified protein under denaturing conditions (9.5 M urea, 2% Nonidet P-40, 5% flmercaptoethanol) was estimated by isoelectric focusing in 12-cm-long horizontal slab gels followed by SDS-PAGE according to the method of O-Farrell [19]. The pH gradient was generated by a mixture of ampholytes (2%) which covered a pI range of p H 3 to 10. To measure the pH profile along the gradient, a standard gel was cut into 5-mm slices, and the ampholytes were eluted in

157

water before the pH was measured. The molecular weight of the native protein was determined by gel filtration on Sephadex G-200. The column (1 cm × 40 cm) was equilibrated at 4°C with 20 mM Pipes, 20 mM fl-mercaptoethanol, 10% (v/v) glycerol and 0.1% (v/v) Tween 20 at pH 7.0 (flow rate: 0.5 ml/h). The calibration standards used were bovine thyroglobulin (670 kDa) bovine Vglobulin (158 kDa), chicken ovalbumin (44 kDa), horse myoglobin (17 kDa) and cyanocobalamin (1.35 kDa). Approximately 200 ng of purified enzyme were applied to the column and it was detected in the effluent by testing for enzyme activity. Protein concentrations were determined according to the method of Bradford [20] with bovine serum albumin as standard. Protein determinaiton had to be performed by silver staining after the Blue Sepharose and folinate-Sepharose columns due to the small amount of protein obtained. A modification of the method described by Ochs was used [21]: photography of silver-stained polyacrylamide gels which contained different concentrations of bovine serum albumin as standard and unknown concentrations of purified protein were developed on cellulose acetate. The relative intensity of the bands was scanned and monitored using a transmittance-reflectance scanning densitometer (Hoefer Scientific Instruments, CA) and areas were calculated with a Mop-3 Integrator (Carl Zeiss, F.R.G.). Results

Purification and properties of the enzyme Methenyltetrahydrofolate synthetase was purified 49000-fold from human liver obtained at

~

b

c

d

e

I

/

i~?~iii~~i!~!~i!~il. . . . . . . ~ i ; ¸~

Fig. 1. SDS-polyacrylamide gel electrophoresis of various fractions in the purification of methenyltetrahydrofolate synthetase. (a) Supernatant; (b) (NH4)2SO 4 precipitate; (c) Blue Sepharose; (d) 5-HCO-HaPteGlu-AH-Sepharose; and (e) molecular weight standards, lysozyme (14400), soybean trypsin inhibitor (21500), carbonic anhydrase (31000), ovalbumin (45000), bovine serum albumin (66200) and phosphorylase b (92 500. The gel was stained with silver nitrate.

TABLE I PURIFICATON OF HUMAN LIVER METHENYLTETRAHYDROFOLATE SYNTHETASE

Supernatant a Precipitation 70% (NH 4) 2SO4 b Blue Sepharose Folinate-Sepharose

Volume (ml)

Total protein (mg)

Total activity (units)

Spec. act. (units/mg)

Purification (-fold)

Yield

110 22 330 18

1910 1 430 0.39 0.02

1.45 1.11 0.93 0.74

7.6" 10 -4 7.8-10 - a 2.4 37

1 1 3160 48680

100 77 64 51

a From 40 g of human liver. b After desalting through G-25 column.

(%)

158

autopsy. A summary of the purification protocol is shown in Table I. The enzyme is extremely labile at 4°C but can be well stabilized with a non-ionic detergent, Tween 20 (0.1% v/v) and glycerol (10% v/v). Under these conditions at pH 7.0, preparations of purified enzyme have been kept for several months at - 8 0 ° C with no substantial loss of activity. At the final purification step, the specific activity at 30°C and pH 6.0 was 37 #mol .rain -1 .mg -1, which corresponds to a turnover number of 1000 min -t. Initial velocities increased linearly in the range 0.1-50 #g enzyme. The enzyme has a single-size subunit with a molecular weight of 27000 according to SDSpolyacrylamide gel electrophoresis (Fig. 1). The molecular weight of the native enzyme was determined to be 27000 by gel filtration on Sephadex G-200. The isoelectric point was estimated to be at pH 7.0 by isoelectric focusing under denaturing conditions. The enzyme shows a very broad pH activity profile, with an apparent maximum at pH 6.5 (Fig. 2). An Arrhenius plot (Fig. 3) shows a linear increase as the temperature is raised from 5°C to 40°C at pH 6.0 and a Q10 value of 1.94 was estimated between 30°C and 40°C. The Arrhenius plot was clearly biphasic, with a slower increase in activity above 40°C and no detectable activity over 60°C, indicating the presence of a rate-limit-

100

Kinetic studies

The K m values of (6R,S)-5-HCO-H4PteGlu, (6S)-5-HCO-H4PteGlu t, (6S)-5-HCO-H4PteGlu 5 and ATP were found to be 4.4 #M, 2.0 #M, 0.6 #M and 20 #M, respectively. Initial-velocity plots (Figs. 4 and 5) intersected, indicating that the enzyme binds substrates in a sequential as opposed to a ping-pong mechanism [22]. Inhibition studies

The ability of various folate analogues to inhibit the activity of the synthetase was determined by varying the concentration of 5-HCO-H4PteGlu t or -Glu 5 in the presence of different fixed concentrations of the inhibitors. Results are shown

o 0

80-

~

ing step in the reaction. Activation energy below 40°C, given by the Arrhenius equation, was estimated to be 11.5 kcal/mol. We attempted to determine the reversibility of the reaction by replacing 5-HCO-H4PteGlu with 5,10-CH+PteGIu and ATP with ADP and inorganic phosphate in the standard assay system. D-Glucose and D-hexose-6-phosphotransferase (EC 2.7.1.1) were also included to trap ATP and regenerate ADP. The assays were performed at pH 6.0 to avoid the rapid transformation of 5,10-CH ÷H4PteGlu to 10-HCO-H4PteGIu and 50 ng of enzyme were used in the system. No 5-HCOHaPteGlu formation could be detected under these conditions.

0

1.2 D

60" D

0.8

0

E 4o E

g

0.6

200

-~ 0.4

o

O'

I

I

5

I

/

6

I

I

7

I

pH

I

8

I

I

9

I

,5

Fig. 2. Activity of methenyltetrahydrofolate synthetase as a function of pH. The initial velocity under varying p H conditions was determined at 3 0 ° C using different buffers in the assay system depending on their p K a. The buffers used, with their p H ranges, were: citrate buffer (pH 4.0-5.5), Mes buffer (pH 5.5-6.5), Pipes buffer (pH 6.5-7.5), Hepes buffer (pH 7.0-8.0) and borate buffer (pH 8.0-10.0).

0.2 0

311

3 II~

313

3114

31i~

76

1/T Fig. 3. Activity of methenyltetrahydrofolate synthetase as a function of temperature. The initial velocities were determined under varying temperature conditions at p H 6.0. On the Arrhenius plot log V values are expressed as 10 s M. m i n - 1 . mg -1 and 1/T as K x 1 0 -3.

159 0.25

0.3-

1/v 0.25

0.2

1/v 0.2

0.15

0.15 0.1

J

/

0.'

/

-0=.2~

-0.1s

-0.05

o.b~

0'.15

1/[_5-CHO H4PteGlu]

0:25

0'35

Fig. 4. Effect of (6R,S)-5-HCO-H4PteGIu I on initial velocity of the synthetase at fixed concentration of ATP. Assay conditions were as in methods and (6R,S)-5-HCO-H4PteGIu concentrations are expressed as #M -1. Symbols on the double reciprocal plot represent ATP at 34 #M (©), 17 #M ( × ) and 8.5/~M (+).

i n T a b l e II. A l l d o u b l e r e c i p r o c a l p l o t s i n t e r c e p t e d on the y-axis, indicating a competitive inhibition pattern. 5-CH3-H4PteGlu, the main folate in hum a n p l a s m a , h a d a K i v a l u e o f 18 # M , w h i l e P t e G l u I ( K i = 55 /xM) a n d H 2 P t e G l u I ( K i = 50 # M ) w e r e less p o t e n t i n h i b i t o r s a n d H 4 P t e G l u h a d no inhibitory effect up to 400 #M. Experiments using 10-HCO-H4PteGlu as a p o t e n t i a l i n h i b i t o r were unsuccessful because of the spontaneous

-0.05

-0.025

0

0.025

0.~05 1/[ATP]

0.075

0'.1

Fig. 5. Effect of ATP on initial velocity of the synthetase at fixed concentration of (6R,S)-5-HCO-H4PteGlu r Assay conditions were as in Methods and ATP concentrations are expressed as #M -1. Symbols on the double reciprocal plot represent (6R,S)-5-HCO-H4PteGIu 1 at 10/~M (©), 5/xM ( × ) and 2.5 #M (+).

transformation of 10-HCO-H4PteGlu to 5,10C H + - H 4 P t e G l u s e e n a t p H 6.0 i n t h e b u f f e r system used. The pentaglutamate derivatives of PteGlu and H2PteGlu were 15-fold more potent inhibitors of the enzyme than their monoglutam a t e f o r m s , w i t h K i v a l u e s o f 3.5 a n d 3.8 # M , respectively. Similarly, methotrexate has no inhibit o r y c a p a c i t y u p t o 350 # M , w h i l e m e t h o t r e x a t e -

TABLE II KINETIC CONSTANTS AND INHIBITION CONSTANTS FOR METHENYLTETRAHYDROFOLATE SYNTHETASE Substrate

Inhibitor

Km a

(/zM) (6R,S)-5-HCO-H4PteGIu 1 (6S)-5-HCO-H 4PteGlu 1 (6S)-5-HCO-H4PteGlu 5

Ki b (#M)

Inhibition pattern

-

4.4

-

-

-

2.0

-

-

-

0.6

-

-

(6R,S)-5-HCO-H4PteGlul (6R,S)-5-HCO-H4PteGlul (6R,S)-5-HCO-H4PteGlu 1 (6R,S)-5-HCO-H4PteGlu 1 (6R,S)-5-HCO-H4PteGlul (6R,S)-5-HCO-H4PteGlul

PteGlu 5 H 2PteGlul H 2PteGlu 5 H4PteGlu 1 5-CH3-H4PteGlul

(6R,S)-5-HCO-H4PteGlul (6R,S)-5-HCO-H4PteGIu 1 (6S)-5-HCO-H4PteGIu 5

methotrexate-Glu I methotrexate-Glu 5 methotrexate-Glu 5

P t e G l u

I

-

55 3.5 50 3.8 > 400 18

competitive competitive competitive competitive no inhibition competitive

-

> 350 15 16

no inhibition competitive competitive

a Km values were determined by double reciprocal ( 1 / V vs. 1/[S]) plots using three different fixed concentrations of ATP at 34/~M,

17 #M and 8.5 ttM. All points represent averages of five measurements. b Ki values were estimated from reports of data ( K i n / V m a x VS. [I]) taken from double reciprocal ( 1 / V vs. 1/[S]) plots using three

different fixed concentrations of inhibitors. Each point represents an average of three measurements.

160 Glu 5 was a competitive inhibitor with a K i value of 15 ~M; the s a m e K i for methotrexate-Glu 5 was obtained whether the competing substrate was (6S)-5-HCO-H 4PteGlu 1 or -Glu 5. Discussion

The present report describes an improved purification procedure for methenyltetrahydrofolate synthetase. The enzyme was purified from human liver using a two-column procedure which combined parts of two recently described purification methods [5,6] and led to a higher overall purification and yield of 49000-fold and 50%, respectively. Human methenyltetrahydrofolate synthetase was purified to homogeneity and appeared as a single band both on SDS-PAGE ( M r'27000) and on isoelectric focusing (pI = 7.0), suggesting that the purified enzyme has a single type of subunit. The only previous reported value for the isoelectric point indicated that it was probably between pH 4.5 and 5.4 [3]. The native enzyme was found to be a monomeric protein by gel filtration with an apparent molecular weight (27000) comparable to those described for bacterial (23000) and rabbit liver (28000) enzymes. Met.henyltetrahydrofolate synthetase was found to have some characteristics similar to those described previously for enzymes isolated from other sources [3-6] but to differ in its affinities for 5-HCO-HaPteGlu and ATP. Furthermore, the human enzyme was found to show specificity for polyglutamate folate substrates. The specific activity at the final purification step was 37 # m o l - m i n -1 .mg -x, which corresponds to a turnover number of 1000 min 1 at 30°C and pH 6.0. The human enzyme seems to be more efficient than the bacterial and rabbit liver synthetases, which had turnover numbers of 88 min 1 and 300 min -1, respectively. The synthetase exhibits activity over a wide pH range with an apparent maximum at pH 6.5. Previous studies with enzyme isolated from sheep liver reported a maximum activity at pH 5.0 with a constant activity between pH 6.5 to 7.5, and rabbit liver enzyme exhibits a broad pH activity profile constant from pH 5.0 to 7.5. Initial-velocity studies suggest that the human

enzyme catalyzes a sequential mechanism as reported for rabbit liver synthetase. The K m of ATP for human methenyltetrahydrofolate synthetase (20 /~M) was significantly different from the reported values for the Lactobacillus casei (1.0/~M) and rabbit liver (0.3 mM) enzymes. The K m values for (6R,S)-5-HCO-H4PteGIu and (6S)-5H C O - H : P t e G l u were found to be 4.4 ttM and 2.0 /~M, respectively, suggesting that the R form is not a substrate or an inhibitor of the synthetase. (6S)-5-HCO-H4PteGlu had a higher g m ( 2 . 0 ~ M ) for the human enzyme compared to those reported for the L. casei (0.6 #M) and rabbit liver (0.5 ~M) enzymes. Since 5-HCO-H4PteGlu is known to be metabolized to polyglutamate derivatives by the enzyme folylpolyglutamate synthetase [23], we determined whether methenyltetrahydrofolate synthetase was able to use the pentaglutamate analogue as a substrate. (6S)-5-HCO-H4PteGIu was found to be a better substrate, with an increased affinity for the enzyme compared to the monoglutamate substrate, with a g m of 0.6 ~tM. This increased affinity is in line with the observation that most pteroyl-polyglutamate substrates bind to their respective enzymes with a greater affinity than the monoglutamate derivatives [23-26]. The lower K m value for (6S)-5-HCOH4PteGlu 5 was not accompanied by any significant change in the apparent Vm~, suggesting that the polyglutamate chain increases substrate binding to the enzyme but does not modify the active site [23]. Similar patterns of K m / V m a x changes for polyglutamate substrates have been observed with mammalian serine hydroxymethyltransferase [27], methylene tetrahydrofolate reductase [28], methylenetetrahydrofolate dehydrogenase [29] and formiminoglutamate:tetrahydrofolate formiminotransferase [29]. Folate and methotrexate polyglutamates are also known to be more potent inhibitors of other folate-dependent enzymes than their monoglutamate forms [23-26,30--32]. This was again borne out for methenyltetrahydrofolate synthetase as P t e G l u 5 ( K i = 3.5/~M), H2PteGlu 5 ( K i = 3.8 ttM) and methotrexate-Glu~ ( g i = 15 ~M) inhibited the enzyme much more potently than folic acid ( g i = 55 /.tM), H2PteGlu ( K i = 50 /LM) or methotrexate ( K i > 350 ttM). As previously described for thymidylate synthase [30], the

161 i n h i b i t i o n c o n s t a n t s for m e t h o t r e x a t e - G l u 5 were the same i n the presence of either the m o n o - or the p e n t a g l u t a m y l folate substrate. The direct i n h i b i t i o n of m e t h e n y l t e t r a h y d r o f o l a t e synthetase b y H 2P t e G l u 5 a n d m e t h o t r e x a t e - G l u 5 could help explain the k n o w n competitive n a t u r e of 5 - H C O H 4 P t e G l u rescue from methotrexate a n d the fact that at high methotrexate c o n c e n t r a t i o n the reduced folate has b e e n shown to be u n a b l e to rescue cells [33,34]. L e u c o v o r i n is often used clinically either as a rescue agent d u r i n g high-dose methotrexate regim e n s [8] or d u r i n g 5-fluorouracil therapy to increase intracellular 5 , 1 0 - C H 2 - H 4 P t e G l u levels [9] a n d thus increase f o l a t e - 5 - F d U M P - t h y m i d y l a t e synthase ternary complex f o r m a t i o n [35]. Consequently, methenyltetrahydrofolate synthetase p o t e n t i a l l y plays a n i m p o r t a n t role in clinical cancer therapy. F u r t h e r studies o n h u m a n tumors are u n d e r way to d e t e r m i n e whether variations in e n z y m e activity could help explain the variable clinical responses o b t a i n e d with t r e a t m e n t regimens including 5-HCO-H4PteGlu.

Acknowledgements W e would like to acknowledge the secretarial work of Mrs. Carole S t - A u b i n a n d the photographic work of Mr. Roger Duclos. This work was s u p p o r t e d b y the Medical Research C o u n c i l of C a n a d a , a n d studentship support was from the C a n c e r Research Society Inc. (Montreal, Canada).

References 1 Greenberg, G.R. (1954) J. Am. Chem. Soc. 76, 1458-1459 2 Peters, J.M. and Greenberg, D.M. (1957) J. Biol. Chem. 226, 329-335 3 Peters, J.M. and Greenberg, D.M. (1958) J. Am. Chem. Soc. 80, 2719-2722 4 Greenberg, D.M., Wynston, L.K. and Nagabhushanam, A. (1965) Biochemistry 4, 1872-1878 5 Grimshaw, C.E., Henderson, G.B., Soppe, G.G., Hansen, G., Mathur, E.J. and Huennekens, F.M. (1984) J. Biol. Chem. 259, 2728-2733 6 Hopkins, S. and Schirch, L.V. (1984) J. Biol. Chem. 259, 5618-5622 7 Bortoluzzi, L. and MacKenzie, R.E. (1983) Can .J. Biochem. Cell Biol. 61,248-253

8 Jolivet, J., Cowan, K.H., Curt, G.A., Clendeninn, N.J. and Chabner, B.A. (1983) N. Engl. J. Med. 309, 1094-1104 9 Machover, D., Schwarzenberg, L., Goldschmidt, E., Tourani, J.M., Michalski, B., Hayat,M., Dorval, T., Misset, J.-L., Jasmin, C., Maral, R. and Mathe, G. (1982) Cancer Treat. Rep. 66, 1803-1807 10 Moran, R.G., Werkheiser, W.C. and Zakrzewski, S.F. (1976) J. Biol. Chem. 251, 3569-3575 11 Coward, J.K., Parameswaran, K.N., Cashmore, A.R. and Bertino, J.R. (1974) Biochemistry 13, 3899-3903 12 Mathews, C.K. and Huennekens, F.M. (1960) J. Biol. Chem. 235, 3304-3308 13 Moran, R.G. and Colman, P.D. (1982) Anal. Biochem. 122, 70-78 14 Rabinowitz, J.C. (1963) Methods Enzymol. 6, 814-815 15 Kaufman, B.T. and Pierce, J.U. (1971) Biochem. Biophys. Res. Commun. 44, 608-613 16 Rabinowitz, J.C. (1960) in The Enzymes (Boyer, P.D., Lardy, H. and Myrback, K., eds.), Vol. 2, pp. 185-252, Academic Press, New York 17 Laemmli, U.K. (1970) Nature (London) 227, 680-685 18 Morrissey, J.H. (1981) Anal. Biochem. 117, 307-310 19 O'FarreU, D.H. (1975) J. Biol. Chem. 250, 4007-4021 20 Bradford, M. (1976) Anal. Biochem. 72, 248 21 Ochs, D.C., McConkey, E.H. and Sommsons, D.W. (1981) Electrophoresis 2, 304-407 22 Segel, I.H. (1975) 'Enzyme Kinetics' Ch. 9, p. 505, John Wiley & Sons, New York 23 McGuire, J.J. and Bertino, J.R. (1981) Mol. Cell. Biochem. 39, 19-48 24 Cichowicz, D., Foo, S.K. and Shane, B. (1981) Mol. Cell. Biochem. 39, 209-228 25 Kisliuk, R.L. (1981) Mol. Cell. Biochem. 39, 331-345 26 McGuire, J.J. and Coward, J.K. (1984) in Chemistry and Biochemistry of Folates, (Blakley, R.L. and Benkovic, S.J., eds.), pp. 135-191, John Wiley & Sons, New York 27 Blakley, R.L. (1957) Biochem. J. 65, 342-348 28 Matthews, R.G. and Baugh, C.M. (1980) Biochem. J. 75, 12-13 29 MacKenzie, R.E. and Baugh, C.M. (1980) Biochim. Biophys. Acta 611, 187-195 30 Allegra,J.C., Chabner, B.A., Drake, C.J., Lutz, R., Rodbard, D. and Jolivet, J. (1985) J. Biol. Chem. 260, 9720-9726 31 Allegra, J.C., Drake, J.C., Jolivet, J. and Chabner, B.A. (1985) Proc. Natl. Acad. Sci. USA 82, 4881-4885 32 Allegra, C.J., Drake, J.C., Jolivet, J. and Chabner, B.A. (1985) in Proceedings of the Second Workshop on Folyl and Antifolyl Polyglutamates (Goldman, I.D., ed.), pp. 194-204, Praeger, New York 33 Pinedo, H.M., Zaharko, D.S., Bull, J.M. and Chabner, B.A. (1976) Cancer Res. 36, 4418-4424 34 Nederbragt, H., Uitendaal, M.P., Van Der Grint, L., Leyua, A. and Pinedo, H.M. (1981) Cancer Res. 41, 1193-1198 35 Houghton, J.A., Maroda, S.J., Phillips, J.O. and Houghton, P.J. (1981) Cancer Res. 41, 144-149